Little correction: I'm not using EQ, only for some cases. I'm mainly balancing the volumes, the Nagras are incredibly dynamic, so that helps having a better listening experience.BDJ wrote: ↑Wed Aug 31, 2022 10:12 am Looks like a project that will keep you busy for a while, and I hope you enjoy it.
I agree with you that many (all?) AI generate tracks lack high frequencies, above 11kHz. With good equipment, and young ears, you can actually hear that. I often also detect a swishy atmosphere, softening of the highs and lows, that I don't like.
Regarding your other 'corrections', I am not sure I would prefer them over the originals (Day by Day, A/B road).
There are no artefacts (pops and clicks) that need removing, and the Nagra recordings are pretty low in tape noise and hiss by themselves; treating them inevitably goes at the expense of some high frequencies. And ADT......ouch!
The ATRAC compression on some tracks is actually there; but it is an excellent compression algorithm, which only results in loss of frequencies above 16kHz. I challenge anyone to a A-B test to hear which track has ATRAC and which one hasn't. The difference is only detected when you inspect the spectrum.
It could be a good thing to normalise volume and EQ a bit, but I would be very careful with it. Twickenham and Apple Studios have very different soundscapes; to what do you normalise the EQ?
My corrections are mainly hum removal and restoring lost spectrum, also since I'm 16, I can hear the difference between lossy and lossless
The "ADT" is not really that, it's a plugin that Cubase has called "MonoToStereo", it's much more enjoyable than a normal ADT if setted correctly